What Is a Virtual Server?
Key Points
- Bradley Knapp explains that a virtual server replicates the four core components of a physical server—CPU, RAM, network, and storage—using software-defined resources.
- Virtual servers are created by partitioning a physical host into “slices,” each receiving a portion of the host’s aggregated compute, memory, network, and storage capacities.
- These slices can vary in size, allowing users to allocate either many small virtual instances or a few large ones depending on workload needs.
- Despite being virtualized, a virtual server (or VSI) behaves to the end‑user just like a traditional physical server, providing the same functional interface and capabilities.
- The discussion emphasizes that not everyone is expected to know these basics upfront, and the video aims to demystify virtual server concepts for newcomers to cloud technology.
Sections
- Introducing Virtual Server Basics - IBM Cloud product manager Bradley Knapp explains what a virtual server is, reassures beginners, and outlines the core physical components—CPU, RAM, storage, and networking—of a traditional server.
- Independent Scaling via Virtual Servers - The speaker explains how virtualization lets each resource—CPU, memory, network, and storage—be scaled independently to match specific workload needs, reducing the inefficiency of oversized physical servers.
- Simplified Virtual Server Overview - The speaker explains that cloud providers handle all underlying infrastructure, allowing users to run a low‑cost virtual server (VSI) with minimal effort, while offering to provide more technical details upon request.
Full Transcript
# What Is a Virtual Server? **Source:** [https://www.youtube.com/watch?v=42fwh_1KP_o](https://www.youtube.com/watch?v=42fwh_1KP_o) **Duration:** 00:07:01 ## Summary - Bradley Knapp explains that a virtual server replicates the four core components of a physical server—CPU, RAM, network, and storage—using software-defined resources. - Virtual servers are created by partitioning a physical host into “slices,” each receiving a portion of the host’s aggregated compute, memory, network, and storage capacities. - These slices can vary in size, allowing users to allocate either many small virtual instances or a few large ones depending on workload needs. - Despite being virtualized, a virtual server (or VSI) behaves to the end‑user just like a traditional physical server, providing the same functional interface and capabilities. - The discussion emphasizes that not everyone is expected to know these basics upfront, and the video aims to demystify virtual server concepts for newcomers to cloud technology. ## Sections - [00:00:00](https://www.youtube.com/watch?v=42fwh_1KP_o&t=0s) **Introducing Virtual Server Basics** - IBM Cloud product manager Bradley Knapp explains what a virtual server is, reassures beginners, and outlines the core physical components—CPU, RAM, storage, and networking—of a traditional server. - [00:03:06](https://www.youtube.com/watch?v=42fwh_1KP_o&t=186s) **Independent Scaling via Virtual Servers** - The speaker explains how virtualization lets each resource—CPU, memory, network, and storage—be scaled independently to match specific workload needs, reducing the inefficiency of oversized physical servers. - [00:06:09](https://www.youtube.com/watch?v=42fwh_1KP_o&t=369s) **Simplified Virtual Server Overview** - The speaker explains that cloud providers handle all underlying infrastructure, allowing users to run a low‑cost virtual server (VSI) with minimal effort, while offering to provide more technical details upon request. ## Full Transcript
Hey there, my name is Bradley Knapp, and I am one of the product managers here at IBM
Cloud and we've come together today to talk about what is a virtual server.
And before we get into the video, I do want to encourage you, please do like and subscribe
so you can get updates whenever we do new videos like this in the future.
But with that, let's go ahead and get in.
What is a virtual server?
You see it everywhere.
Everyone talks about it.
And if you don't know, that's OK, because the tech press, as a rule, makes an assumption
that everyone has a baseline of knowledge that is as old as theirs is.
And the folks that have been in the tech press have been there for at least 10 years.
They've been there for 15 years.
And so, when you start out in your career in tech, you don't know the answers to all
these questions.
And there's nothing wrong with that.
It's a perfectly legitimate question and it's one that we want to help you out with.
So, if we think about hardware in a data center, in your house, in your car, but physical hardware,
as a general rule, a computer is a box, right.
It's a physical box.
And inside that physical box, you're going to have one or more CPUs, that does compute.
You're going to have some amount of RAM that does storage - I guess we could label these
CPU, RAM - and that storage is directly attached to the CPU, the RAM is attach to the CPU,
so it's very close.
It's for the actual operation of the machine.
And then you're going to have network, right?
Because we're talking about virtual servers in cloud.
So, network is important.
That is how the box talks to everything else.
And then you're going to have traditional storage.
Want to put traditional storage over here for a moment because traditional storage can
go inside the server, or it can be out on its own.
But either way, those are kind of your four fundamental building blocks.
So, a virtual server is a virtual compute, virtual RAM, virtual network, virtual storage,
but it's presented in a way that it acts like a server does.
So, if we consider a physical server to be a physical device, you can put your hands
around, what a virtual server does is it takes this physical host and it divides it up into
slices.
Right.
You've got slice one, slice two, slice three, slice four.
Your slices take all of the aggregate resources on the physical server, and they present just
a little bit.
Now, you don't have to do it that way.
If instead of these a little bitty slices, you want a great big, huge slice and then
you want to fill the box up with a little slice at the end, that's fine, too.
But at the end of the day, a virtual server is part of a physical server that you get
access to.
And inside that virtual server, or that VSI, you are going to have compute, you're going
to have RAM, you're going to have network, and you're going to have storage.
And it is presented to you, to the user, to the consumer, just like a physical box would
be.
But because it's virtualized, that virtualization, what that means is that we're taking and dividing
these up, we can scale these resources up and down.
So, if you need a lot of compute in just a little bit of RAM, that's fine, there's no
problem with that.
We can build the virtual server that way.
If you need a little bit of compute in a little bit of RAM, but a lot of network, right.
Like it's running a website where you're constantly sending data out.
That's fine.
We can scale a virtual server that way.
Each of these metrics is independently scalable.
So, you can have a little bit of compute, a little bit of RAM, a little bit of network
and a ton of storage.
That's fine because, again, a virtual server is designed to solve whatever your specific
needs are, not the needs of the people who build the box.
And so why is this important?
Well, as time has gone on and very specifically over the last 10 years, our physical server
boxes, right.
The physical boxes that go into the data center have gotten more and more and more powerful.
And that power is great, but single workloads can't take advantage of it.
So, using my example before, just a normal basic Web server, right, a normal basic web
server might need 4 VCPUs, 4 CPU's, it might need 64 GBs of RAM, it might need a 1 Gbit
network connection and it might need, I don't know, 250 GBs of storage.
Well, a physical server that meets those requirements takes up a lot of space in a data center.
Right.
Because our data center, if you go in there and you've got a data center rack, right,
I've got my rack here, and the racks have shelves, and the shelves are 1U tall, 2U tall,
4U tall, 8U tall, you know.
The smallest unit is a 1U.
Now, there are 1/2-U and you can take 1U and make it into 3 servers, but just for our purposes,
let's call it a physical slot in the data center rack, 1U.
Well, this virtual server, if we were to put it in the physical server, would need that
whole U of space.
That's not space efficient, it's power efficient, it's not floor print efficient, and it's bad
for general efficiency, not just of compute, but it's not great for the environment either.
So, we want to drive density.
So, what you do is you take that physical box that takes up that same 1U of space, right?
So, 1U.
So, we're going to have a physical box that takes up 1U of space, but we're going to make
it super powerful so that it can hold 50 of these virtual servers because we have the
ability to do that.
With the advances in computing technology, we can pack 50 virtual servers of our little
Web server into 1U of space.
This is a great thing.
It drives down the cost because these servers, if I can put 50 of them on one physical box,
the cost for my physical box over here, I divide up 50 ways, makes it much less expensive
for you.
And it makes the compute resources that previously you'd have to go out in a data center space,
get the power into it, bring the network in.
Now, you don't have to do that anymore.
The cloud provider takes care of all of that for you.
And all you care about is that you're going to have your one little VSI over here that's
going to cost you 20 cents an hour.
So that's the beauty of a virtual server.
That's how they work.
Now, obviously, we didn't get into the great fine technical details.
If you'd like to know more about that, please do drop us a note.
Let us know.
We're happy to give you more information.
But that's a virtual server.
That's how they work and that's why they're important.